Urbanization impacts species in many different ways, including behavior, morphology, and physiology. Although many of the documented phenotypic shifts show evidence of a genetic basis, the actual genetic differences and the genes they occur in are unknown for most urban adaptations. In fact, it seems we know very little at the genetic level regarding which genes are under selection in urban environments. Molecular methods are rapidly advancing, providing new powerful approaches for detecting selection at the genetic level and identifying candidate genes for local adaptation.
Some researchers have taken an approach of sequencing specific genes related to specific traits they know to differ in urban populations. For example, Mueller et al. (2013) targeted several genes related to exploratory behavior, circadian rhythm, migration, stress, harm avoidance, and aggression in the common blackbird (Turdus merula) based on evidence from previous research that birds had shifted circadian rhythms, reduced stress responses, and reduced flight-initiation distances. Using a candidate gene approach in which they sequenced microsatellites and SNPs within target genes, they detected differences in DRD4 and SERT in urban birds. These genes are related to dopamine and serotonin, which play a major role stress responses and exploratory behavior. While this approach benefits from a targeted approach (which means results are more easily interpretable), it also requires identification of the genes that may be underlying variation before sequencing.
Others have taken a different approach, with reduced representation sequencing, such as RADseq, or whole genome / transcriptome sequencing. These approaches are beneficial because they do not rely on a priori identifications of genes and are more likely to detect genetic changes that we may not necessarily expect. But they come with downsides, not least of which is the high cost of whole genome sequencing. In addition, results may be difficult to interpret with SNPs falling in regions that are either not annotated or have ambiguous or broad regulatory functions. For example, Harris and Munshi-South sequenced the transcriptomes of white footed mice (Peromyscus leucopus) and identified a number of genetic regions that differ between urban and non-urban populations in New York City (Harris et al. 2013, 2017). Harris et al. (2017) found 19 different outlier loci, and some of these are associated with metabolic and immune functions, which are very probably under differential selection in city versus rural populations. But many of the identified regions are of uncertain function or of ambiguous importance.
A more targeted sequencing method is exome capture, which can either target the entire exome (i.e., all expressed genes) or a targeted subset of the exome related to functions of interest. I’m aware of only a couple of people doing this in the urban system – myself with anole lizards, and Jason Munshi-South with white-footed mice. (Let us know if you are doing this – I would love to chat with you about this!) But others have found success in this method in measuring selection and local adaptation in other systems (e.g., Roffler et al. 2016). The benefits of this approach is that any SNPs detected lie within expressed genes, so interpretation may be easier than with RADseq approaches, and depending on the bait tiling approach (e.g., randomly distributed baits or full tiling of exons) and target (specific genes/exons or entire exome), likelihood of detecting local genetic variation may be greater than with other reduced representation approaches.
Despite this wealth of molecular resources at our disposal, I’ve been shocked to find how few gene regions we actually know to differ between urban and non-urban individuals. I am compiling a list of all of those I could find and I would like your help to add to this list – it seems like I must be missing quite a few! Ultimately, by building this list we can start to find commonalities between species responses to urbanization at the genetic level, which will create new research ideas and opportunities for us all. If you know of any examples other than the ones listed below, please comment here and I will add them to the list!
- White-footed Mouse (Peromyscus leucopus) – metabolic, immune, unknown functions (Harris et al. 2013, Harris and Munshi-South 2017)
- Proteasome 26S subunit, non-ATPase, 9
- Transmembrane 9 superfamily member 1
- Tubulin folding cofactor E-like
- OTU domain containing 3
- X-ray repair complementing defective repair in CHC3
- A kinase (PRKA) anchor protein 8
- Autophagy-related 2A
- GRAM domain containing 3
- Xanthine dehydrogenase
- Cytochrome c oxidase III
- Glyoxylate reductase/hydroxypyruvate reductase
- Serine peptidase inhibitor, clade A, member 3N
- GTPase, very large interferon inducible 1
- Oculocerebrorenal syndrome of Lowe
- NADH dehydrogenase 1 alpha subcomplex 11
- Translocation protein SEC62
- 39S ribosomal protein L51
- Histone H1-like protein in spermatids 1
- PHD finger protein 8
- Aldo-keto reductase family 1, member C12
- Camello-like 1
- Cytochrome P450 2A15
- Fibrinogen alpha chain
- Solute carrier organic anion transporter family member 1A5
- Serine protease inhibitor a3c
- Alpha-1-acid glycoprotein 1
- Birds (Cygnus atratus, Turdus merula, Parus major) – stress response, exploratory behavior (van Dongen et al. 2015, Mueller et al. 2013, Riyahi et al. 2015)
- SERT
- DRD4
- Mosquito (Anopheles gambiae) – DDT resistance (Fossog Tene et al. 2013)
- detoxification enzymes and ABC transporters; CYP6M2 and GSTD1-6
- Bumblebee (Bombus terrestris) – 56 loci (Theodorou et al. 2018)
- Killifish (Fundulus heteroclitus) – PCB / toxin exposure (Whitehead et al. 2012)
- genes in AHR-mediated signalling pathway
- White clover (Trifolium repens) – hydrogen cyanide production (Johnson et al. 2018)
- New Lit Alert: Urban environment determines population genetics in the green toad, Bufotes viridis - September 11, 2023
- Contribute to the New York Canid Project! - July 21, 2023
- Parallel Urban Adaptation from Phenotype to Genotype in Anolis Lizards - January 19, 2023